Qwen2.5-0.5B-Open-R1-Code-GRPO / all_results.json
rasdani's picture
Model save
b6cdafc verified
{
"total_flos": 0.0,
"train_loss": 0.05661330611514859,
"train_runtime": 3774.5693,
"train_samples": 27839,
"train_samples_per_second": 2.119,
"train_steps_per_second": 0.132
}